# US 12,223,254 B2 · Using machine learning to predict performance of secure documents > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,223,254 B2 - Document ID: US12223254B2 - Status: Granted - Kind: B2 - Application number: 18/476,060 - Filed: 2023-09-27 - Published: 2024-01-18 - Granted: 2025-02-11 - Earliest claimed: 2020-10-09 - Assignees: DocuSign, Inc. - Applicants: DocuSign, Inc. - Inventors: Soroush Salehian; William Pittman; John Barcellos; Santiago Szuchmacher; Chris Marshall; Jeong Woo Chang; Dylan Ray Häs - Web page: https://patentgrep.com/patent/US12223254B2 ## CPC - G06F 21/64: Protecting data integrity, e.g. using checksums, certificates or signatures - G06F 40/106: Display of layout of documents; Previewing - G06F 18/2193: {based on specific statistical tests} - G06F 21/6209: {to a single file or object, e.g. in a secure envelope, encrypted and accessed using a key, or with access control rules appended to the object itself} - G06F 40/186: Templates - G06F 40/216: using statistical methods - G06N 20/00: Machine learning - G06V 30/40: Document-oriented image-based pattern recognition - G06F 40/174: Form filling; Merging - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A system and a method are disclosed for detecting that a requesting user is transmitting a request to a recipient user to perform one or more tasks with respect to a secure document. The system extracts features associated with the secure document and inputs the features into a machine learning model that outputs one or more probabilities corresponding to a potential failure of the request. Based on the one or more probabilities, the system generates one or more feature modification activities to improve a likelihood of the recipient user completing the request. The system provides the one or more feature modification activities to at least one of the requesting user and the recipient user. ## Claims ### Claim 1 A computer-implemented method, comprising: detecting, using at least one processor, a transmission of a request from a requesting computing device to a recipient computing device to perform one or more tasks associated with a secure document; identifying, using the at least one processor, based on the request in the detected transmission, one or more machine learned models in a plurality of machine learned models, the one or more machine learned models have been trained on one or more features of one or more historical secure documents and task completion data associated with the one or more historical secure documents, and determining, using the identified one or more machine learned models, one or more probabilities associated with a failure of the recipient computing device to complete performance of the one or more tasks, wherein the determined one or more probabilities are indicative of a failure to complete performance of the one or more tasks based on at least one of the following: a type of task in the one or more tasks, a time of transmission of the request, a time associated with performance of at least one task in the one or more tasks, and any combination thereof; generating, using at least one processor, based on the determined one or more probabilities, one or more feature modification recommendations, each feature modification recommendation in the one or more feature modification recommendations being associated with a value of probability indicative of an improved likelihood of a successful completion of performance of the one or more tasks by the recipient computing device; and presenting, using at least one processor, the generated one or more feature modification recommendations and associated respective values of probability on a graphical user interface of the requesting computing device. ### Claim 2 (depends on 1) The method of claim 1, wherein the generating includes comparing each value of probability to one or more threshold values of probability; and selecting, based on the comparing, at least one feature modification recommendation associated with a value of probability exceeding the one or more threshold values of probability for presentation of the at least one feature modification recommendation on the graphical user interface. ### Claim 3 (depends on 1) The method of claim 1, wherein the identifying includes identifying two or more machine learned models in the plurality of machine learned models to simultaneously determine at least one of: a plurality of probabilities associated with a failure to complete performance of the one or more tasks, one or more variations of the one or more probabilities associated with a failure to complete performance of the one or more tasks, and any combination thereof. ### Claim 4 (depends on 1) The method of claim 1, wherein the one or more feature modification recommendations include a threshold number of feature modification recommendations being associated with a value of probability indicative of an improved likelihood of a successful completion of performance of the one or more tasks. ### Claim 5 (depends on 4) The method of claim 4, wherein the one or more feature modification recommendations include one or more feature modification recommendations to modify one or more features associated with the secure document, the one or more feature modification recommendations including an updated value of probability indicative of the improved likelihood of the successful completion of performance of the one or more tasks in response to modifying the feature. ### Claim 6 (depends on 1) The method of claim 1, wherein the one or more tasks include at least one of: signing of the secure document, uploading of an attachment to the secure document, adding information into an input field of the secure document, approving the secure document, and any combination thereof. ### Claim 7 (depends on 1) The method of claim 1, wherein the one or more feature modification recommendations include at least one of the following: removing at least one task from the request, changing a frequency of reminders to complete performance of the one or more tasks, changing a layout of one or more input fields in the secure document, delaying a transmission of the request, and any combination thereof. ### Claim 8 A system, comprising: at least one processing circuitry; and at least one non-transitory storage media storing instructions, that when executed by the at least one processing circuitry, cause the at least processing circuitry to: detect a transmission of a request from a requesting computing device to a recipient computing device to perform one or more tasks associated with a secure document; identify, based on the request in the detected transmission, one or more machine learned models in a plurality of machine learned models, the one or more machine learned models have been trained on one or more features of one or more historical secure documents and task completion data associated with the one or more historical secure documents, and determine, using the identified one or more machine learned models, one or more probabilities associated with a failure of the recipient computing device to complete performance of one or more tasks associated with the secure document, wherein the determined one or more probabilities are indicative of a failure to complete performance of the one or more tasks based on at least one of the following: a type of task in the one or more tasks, a time of transmission of the request, a time associated with performance of at least one task in the one or more tasks, and any combination thereof; generate, based on the determined one or more probabilities, one or more feature modification recommendations, each feature modification recommendation in the one or more feature modification recommendations being associated with a value of probability indicative of an improved likelihood of a successful completion of performance of the one or more tasks by the recipient computing device, wherein generation of the one or more feature modification recommendations includes: comparing each value of probability to one or more threshold values of probability; and selecting, based on the comparing, at least one feature modification recommendation associated with a value of probability exceeding the one or more threshold values of probability for presentation of the at least one feature modification recommendation on a graphical user interface of the requesting computing device; and present the generated one or more feature modification recommendations and associated respective values of probability on the graphical user interface of the requesting computing device. ### Claim 9 (depends on 8) The system of claim 8, wherein identification of the one or more machine learned models includes detecting transmission of a request to perform the one or more tasks associated with the secure document. ### Claim 10 (depends on 9) The system of claim 9, wherein the one or more feature modification recommendations include at least one of the following: removing at least one task from the request, changing a frequency of reminders to complete performance of the one or more tasks, changing a layout of one or more input fields in the secure document, delaying a transmission of the request, and any combination thereof. ### Claim 11 (depends on 8) The system of claim 8, wherein identification of the one or more machine learned models includes identifying two or more machine learned models in the plurality of machine learned models to simultaneously determine at least one of: a plurality of probabilities associated with a failure to complete performance of the one or more tasks, one or more variations of the one or more probabilities associated with a failure to complete performance of the one or more tasks, and any combination thereof. ### Claim 12 (depends on 8) The system of claim 8, wherein the one or more feature modification recommendations include a threshold number of feature modification recommendations being associated with a value of probability indicative of an improved likelihood of a successful completion of performance of the one or more tasks; wherein the one or more feature modification recommendations include one or more feature modification recommendations to modify one or more features associated with the secure document, the one or more feature modification recommendations including an updated value of probability indicative of the improved likelihood of the successful completion of performance of the one or more tasks in response to modifying the feature. ### Claim 13 (depends on 8) The system of claim 8, wherein the one or more tasks include at least one of: signing of the secure document, uploading of an attachment to the secure document, adding information into an input field of the secure document, approving the secure document, and any combination thereof. ### Claim 14 A computer program product comprising a non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processing circuitry, cause the at least one programmable processing circuitry to: detect a transmission of a request from a requesting computing device to a recipient computing device to perform one or more tasks associated with a secure document; identify, based on the request in the detected transmission, one or more machine learned models in a plurality of machine learned models, the one or more machine learned models have been trained on one or more features of one or more historical secure documents and task completion data associated with the one or more historical secure documents, and determine, using the identified one or more machine learned models, one or more probabilities associated with a failure of the recipient computing device to complete performance of the one or more tasks, wherein the determined one or more probabilities are indicative of a failure to complete performance of the one or more tasks based on at least one of the following: a type of task in the one or more tasks, a time of transmission of the request, a time associated with performance of at least one task in the one or more tasks, and any combination thereof; generate, based on the determined one or more probabilities, one or more feature modification recommendations, each feature modification recommendation in the one or more feature modification recommendations being associated with a value of probability indicative of an improved likelihood of a successful completion of performance of the one or more tasks by the recipient computing device; and present the generated one or more feature modification recommendations and associated respective values of probability on a graphical user interface of the requesting computing device. ### Claim 15 (depends on 14) The computer program product of claim 14, wherein generation of the one or more feature modification recommendations includes comparing each value of probability to one or more threshold values of probability; and selecting, based on the comparing, at least one feature modification recommendation associated with a value of probability exceeding the one or more threshold values of probability for presentation of the at least one feature modification recommendation on the graphical user interface.